CN113538041B - Power package recommendation method and device based on load curve clustering analysis - Google Patents
Power package recommendation method and device based on load curve clustering analysis Download PDFInfo
- Publication number
- CN113538041B CN113538041B CN202110730352.2A CN202110730352A CN113538041B CN 113538041 B CN113538041 B CN 113538041B CN 202110730352 A CN202110730352 A CN 202110730352A CN 113538041 B CN113538041 B CN 113538041B
- Authority
- CN
- China
- Prior art keywords
- power
- electricity
- representing
- price
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The embodiment of the invention discloses a power package recommendation method based on load curve clustering analysis, which comprises the following steps: acquiring historical electricity utilization data of terminal users, and analyzing a typical electricity utilization mode of each terminal user by using a DBSCAN clustering algorithm; optimizing the combination in the typical electricity utilization mode of the terminal user according to a time-of-use electricity price structure and an electricity price level by using a mixed integer nonlinear programming method, and determining an optimal electricity utilization mode package; recommending the optimal power consumption mode package for each terminal user to the terminal user. The method can effectively reduce the electricity cost of the terminal user while ensuring the return rate of the retailer under the constraint of considering the elastic constraint of the power demand and the operation of the power distribution network, and improves the competitiveness of the retailer in the market.
Description
Technical Field
The embodiment of the invention belongs to the field of electricity, and particularly relates to a method and a device for recommending a power package based on load curve clustering analysis.
Background
As the electricity market has developed, retail electricity prices have evolved from fixed, uniform pricing to dynamic, even real-time pricing. For ordinary home users, however, this means that they will be directly faced with the risk of price fluctuations, which is unacceptable. Therefore, the time-of-use electricity prices between the fixed price and the dynamic price are widely adopted on a global scale. However, when the power retailer makes a price package of the retail market, the power retailer is often optimized only for the price level of the time-of-use electricity price based on the overall behavior of the market, so as to maximize the income of the power retailer. However, such retail price packages do not take into account the different electricity usage patterns of the different end users on the one hand, nor the ladder structure of the time of use prices, thereby neglecting the complementarity between the different end users on the time scale.
Disclosure of Invention
In order to solve the above technical problems, embodiments of the present invention provide a method and an apparatus for recommending a power package based on load curve clustering analysis:
a power package recommendation method based on load curve clustering analysis comprises the following steps:
obtaining historical electricity utilization data of terminal users, and analyzing a typical electricity utilization mode of each terminal user by using a DBSCAN clustering algorithm;
optimizing the combination in the typical electricity utilization mode of the terminal user according to a time-of-use electricity price structure and an electricity price level by using a mixed integer nonlinear programming method, and determining an optimal electricity utilization mode package;
recommending the optimal power consumption mode package for each terminal user to the terminal user.
A power package recommendation device based on load curve cluster analysis comprises:
the system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring historical electricity consumption data of terminal users and analyzing a typical electricity consumption mode of each terminal user by using a DBSCAN clustering algorithm;
the processing module is used for optimizing the combination in the typical electricity utilization mode of the terminal user according to the time-of-use electricity price structure and the electricity price level by using a mixed integer nonlinear programming method, and determining an optimal electricity utilization mode package;
and the execution module is used for recommending the optimal power consumption mode package aiming at each terminal user to the terminal user.
The embodiment of the invention has the beneficial effects that: analyzing a typical power utilization mode based on a DBSCAN clustering algorithm and historical power utilization data of the terminal users, and determining the power utilization mode of each terminal user; and for different power utilization modes, optimizing the stepped structure and the power price level of the time-of-use power price by using a mixed integer nonlinear programming method so as to minimize the power utilization cost of the terminal user under the constraint of ensuring the return rate of retailers and recommend the package with the minimum power utilization cost to the user. Under the constraint of considering the elastic constraint of power demand and the constraint of power distribution network operation, the method can effectively reduce the power consumption cost of the terminal user while ensuring the return rate of the retailer, and improve the competitiveness of the retailer in the market. The method can be applied to transaction service charging analysis of the future electric power retail market by combining terminal user clustering and power utilization pattern analysis, and assists the electric power retail market to formulate retail price packages. The invention learns the user load composition and the power utilization behavior rule through data mining, and can be applied to customer management strategy formulation, power selling decision optimization and customized service differentiation of power selling companies.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a power package recommendation method based on load curve cluster analysis according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an electrical configuration provided by an embodiment of the present invention;
FIG. 3 is a schematic diagram provided by an embodiment of the present invention;
FIG. 4 is another schematic diagram provided in accordance with an embodiment of the present invention;
FIG. 5 provides another schematic representation of an embodiment of the present invention;
fig. 6 is a block diagram of a basic structure of an electric power package recommendation device based on load curve cluster analysis according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.
In some of the flows described in the present specification and claims and in the above figures, a number of operations are included that occur in a particular order, but it should be clearly understood that these operations may be performed out of order or in parallel as they occur herein, with the order of the operations being indicated as 101, 102, etc. merely to distinguish between the various operations, and the order of the operations by themselves does not represent any order of performance. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor limit the types of "first" and "second" to be different.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a diagram illustrating a method for recommending a power package based on load curve clustering analysis according to an embodiment of the present invention, where the method specifically includes the following steps:
s110, obtaining historical electricity utilization data of terminal users, and analyzing a typical electricity utilization mode of each terminal user by using a DBSCAN clustering algorithm;
s120, optimizing the combination in the typical electricity utilization mode of the terminal user according to a time-of-use electricity price structure and an electricity price level by using a mixed integer nonlinear programming method, and determining an optimal electricity utilization mode package;
and S130, recommending the optimal power utilization mode package for each terminal user to the terminal user.
The electric power package recommending device based on load curve clustering analysis in the embodiment of the invention analyzes a typical power consumption mode based on a DBSCAN clustering algorithm and historical power consumption data of terminal users, and determines the power consumption mode of each terminal user; and for different power utilization modes, optimizing the stepped structure and the power price level of the time-of-use power price by using a mixed integer nonlinear programming method so as to minimize the power utilization cost of the terminal user under the constraint of ensuring the return rate of retailers and recommend the package with the minimum power utilization cost to the user. Under the constraint of considering the elastic constraint of power demand and the constraint of power distribution network operation, the method can effectively reduce the power consumption cost of the terminal user while ensuring the return rate of the retailer, and improve the competitiveness of the retailer in the market. By combining the clustering of the terminal users and the analysis of the power utilization mode, the method can be applied to transaction service charge analysis of the future power retail market and assist the power retail market in formulating retail price packages. The invention learns the user load composition and the power utilization behavior rule through data mining, and can be applied to customer management strategy formulation, power selling decision optimization and customized service differentiation of power selling companies.
The embodiment of the invention provides a method for calculating the power utilization mode of the terminal user by utilizing a DBSCAN clustering algorithm formulaComprises the following steps:
wherein the vectorRepresents the electricity consumption vector of the terminal user j in tau days, n (-) represents the standardization operation,denotes a ratio of power usage by said end user j for a time period t times t days over the user's total day power usage on that day, H denotes a time period of historical power usage data, TPS denotes a set of typical power usage patterns, TP denotes a typical power usage pattern for an end user, | | 2 Representing the L2 norm of the vector.
The embodiment of the invention provides a method for optimizing the power utilization mode of a terminal user according to a time-of-use power price structure and a power price level by using an optimized objective function of mixed integer nonlinear programming;
the optimization objective function is: minS tp =∑ j∈J ∑ t∈T E(L j,t )·f j,t (r j,t )·r j,t
Wherein S is tp Representing a total electricity usage cost for the end user; j represents the set of all end users; t represents a decision period; e (L) j,t ) Representing a desire of the end user j to use electricity during a t-th time period of the day;r j,t represents the electricity price of the end user j in the t-th time period of each day, f j,t (. Cndot.) represents the elasticity of the end user j's power demand with respect to real-time electricity prices at the t-th time period of each day.
The embodiment of the invention provides a method for calculating the power utilization mode of a terminal user by using a DBSCAN clustering algorithm formula, which comprises the following steps:
step 6, if the current cluster core object queueThen the current cluster C is clustered k After the generation is finished, the typical electricity utilization mode set is updatedUpdating the set of core objects Ω = Ω -C k Turning toEntering a step 5; if the current cluster core object queueThen the core object set Ω = Ω -C is updated k ,
And 8, outputting and initializing a typical power consumption mode set TPS.
When the power consumption mode of the terminal user is optimized according to the time-of-use power price structure and the power price level, modeling is carried out on the power purchase cost, the risk price premium and the expected retail income according to the input expected return rate, the time-of-use power price step number, the daily decision period, the number and price for signing the long-term contract, the power price elastic equation coefficient, the risk weighting factor, the risk price and the spot price expectation, so that a power purchase cost model, a risk price premium model and an expected retail income model are obtained, and a constraint model is constructed, so that a time-of-use power price constraint model, a power demand elastic constraint model and a power distribution network operation constraint model are obtained.
The embodiment of the invention provides a method for optimizing a power utilization mode of a terminal user according to a time-of-use power price structure and a power price level by using an optimization objective function of mixed integer nonlinear programming, wherein the power utilization mode of the terminal user is optimized according to the time-of-use power price structure and the power price level by using a time-of-use power price constraint model, a power demand elastic constraint model and a power distribution network operation constraint model;
wherein the time-of-use electricity price constraint model is as follows:
y i,j =(y i,1,j ,y i,2,j ,…,y i,t,j ,…,y i,|T|,j )
wherein N is pb A step number representing a time of use electricity price for the retail price package; y is i,t,j Is a boolean variable whose value is 1 or 0, indicating whether the ith electricity price block covers t of the day, if so, y i,t,j =1, otherwise y i,t,j =0;Is the electricity purchase cost of the retailer for customer j at i, including costs from forward contracts and spot markets;is a risk premium for user j for the retailer.
The elastic constraint model of the power demand is as follows:
wherein r is 0,t Is a nominal retail price; beta is a 0,j ,β 0,j And beta 0,j Are parameters.
The power distribution network operation constraint model is as follows:
U i,j =(u i,1,j ,u i,2,j ,…,u i,|T|,j )
V i,j =(v i,1,j ,v i,2,j ,…,v i,|T|,j )
where ρ is l,k Is the power distribution coefficient, representing the relative change in active power that occurs on line l due to the actual power change at node k;is the power limit of line l; k and L are respectively a node set and a line set of the power distribution network; u shape i,j And V i,j Are binary vectors whose elements are 0 or 1.
The embodiment of the invention provides a method for determining an optimal power consumption mode package, which comprises the following steps:
determining the optimal power utilization pattern package by utilizing a preset end user power utilization expectation model, a retailer power purchase cost model, a risk price premium model in the retail price and an expected retail income model of the retailer;
wherein the end user's electricity usage expectation model:
wherein the content of the first and second substances,the t-th element of the power usage pattern of end user j.
The electricity purchase cost model of the retailer is as follows:
L j,t =n(E(L j,t ))·E(Q j )
wherein N is F Is the number of forward contracts the retailer has signed;andrespectively indicating the number and price level of the retailer's contracted forward contracts m; beta is a fc Is a weighted factor, β, between the expected revenue and profit risk of the retailer fc ∈[0,+∞);Representing retail risks arising from differences between the forward contract and the end user load expectations; alpha is used to calculate the conditional value-at-risk, CVaR) value-at-risk (VaR); β is a given confidence level; n is a radical of S Represents the number of samples;is the cost of electricity purchased by the retailer from the spot market;is a binary variable, ifRepresenting a lead time belonging to a forward contract m at t;is the spot price at t; e (L) j,t ) The desire of the time end user j to use electricity at time t; e (Q) j ) Indicating the end user j's desire for daily electricity usage.
A risk premium model in the retail price:
the expected retail revenue model for the retailer:
wherein W represents the retail profit of the retailer; e represents the profitability of the retailer's demand.
The embodiment of the invention also comprises the following steps:
linearize retail risk due to differences between the forward contract and the end user load expectations:
M j,n ≥0
Referring to fig. 2, the power distribution network has 37 network nodes in total. The system has 31 terminal users, 5 feeders, and the data and parameters of the feedback feeder are shown in table 1. In an embodiment of the present invention, referring to fig. 3, the L2 norm distance value between samples of the normalized end-user historical electricity consumption data is mainly distributed between 0 and 0.2. The historical electricity consumption data contains 2790 daily electricity consumption data for a total of 31 end users for 90 days.
TABLE 1. Mini Engine parameters
By selecting epsilon =0.2 and minpts =300 as the domain parameters, 7 typical power usage patterns can be obtained by the method of the present invention, please refer to fig. 4. The load peaks for typical power usage patterns 1 and 3 occur at 20:00 to 21:00, the peak load of typical power usage pattern 4 occurs in the early morning of each day, typical power usage patterns 2, 5, 7 are all load patterns that include two peak load periods, and the two peak load periods are located early in the morning and late afternoon, respectively, and the total daily power usage of typical power usage pattern 6 remains relatively stable. See fig. 5 for an example of a retail price package for an end user obtained by the method of the present invention.
As shown in fig. 6, to solve the above problem, an embodiment of the present invention further provides an electric power package recommendation apparatus based on load curve cluster analysis, including: the system comprises a fetching module 2100, a processing module 2200 and an executing module 2300, wherein the fetching module 2100 is used for obtaining historical electricity utilization data of the terminal users and analyzing a typical electricity utilization mode of each terminal user by using a DBSCAN clustering algorithm; a processing module 2200, configured to optimize a combination in a typical electricity consumption mode of the end user according to a time-of-use electricity price structure and an electricity price level by using a mixed integer nonlinear programming method, and determine an optimal electricity consumption mode package; an executing module 2300, configured to recommend the optimal power usage pattern package for each end user to the end user.
The electric power package recommending device based on load curve clustering analysis in the embodiment of the invention analyzes a typical power consumption mode based on a DBSCAN clustering algorithm and historical power consumption data of terminal users, and determines the power consumption mode of each terminal user; for different power utilization modes, a mixed integer nonlinear programming method is utilized to optimize the stepped structure and the power price level of the time-of-use power price, so that the power utilization cost of the terminal user is minimized under the constraint of ensuring the return rate of a retailer, and a package with the minimum power utilization cost is recommended to the user. Under the constraint of considering the elastic constraint of power demand and the constraint of power distribution network operation, the method can effectively reduce the power consumption cost of the terminal user while ensuring the return rate of the retailer, and improve the competitiveness of the retailer in the market. By combining the clustering of the terminal users and the analysis of the power utilization mode, the method can be applied to transaction service charge analysis of the future power retail market and assist the power retail market in formulating retail price packages. The invention learns the user load composition and the power utilization behavior rule through data mining, and can be applied to customer management strategy formulation, power selling decision optimization and customized service differentiation of power selling companies.
In some embodiments, the processing module 2100 is configured to calculate the power usage pattern of the end user using a DBSCAN clustering algorithm formulaComprises the following steps:
wherein the vectorRepresents the electricity consumption vector of the end user j in tau days, n (-) represents the normalization operation,denotes a ratio of power usage by said end user j for a t-th time period on a t day to the power usage of the user all day on that day, h denotes a time period of historical power usage data, TPS denotes a set of typical power usage patterns, TP denotes a typical power usage pattern for an end user, | · | 2 Representing the L2 norm of the vector.
In some embodiments, the execution module 1300 is configured to optimize the end user's power usage pattern in terms of a time of use power rate structure and a power rate level using a mixed integer nonlinear programming optimization objective function;
the optimization objective function is: minS tp =∑ j∈J ∑ t∈T E(L j,t )·f j,t (r j,t )·r j,t
Wherein S is tp Representing a total electricity cost for the end user; j represents the set of all end users; t represents a decision period; e (L) j,t ) Representing a desire of the end user j to use electricity during a t-th time period of the day; r is a radical of hydrogen j,t Represents the electricity price of the end user j in the t-th time period of each day, f j,t (. H) represents the elasticity of the end user j's electricity demand with respect to real-time electricity prices at the tth time period of the day.
In some embodiments, the processing module 2200 is configured to calculate the electricity usage pattern of the end user by using the DBSCAN clustering algorithm formula, and includes:
step 6, if the current cluster core object queueThen the current cluster C is clustered k After the generation is finished, the typical electricity consumption mode set is updatedUpdating the set of core objects Ω = Ω -C k Turning to step 5; if the current cluster core object queueThen the core object set Ω = Ω -C is updated k ,
And 8, outputting and initializing a typical power utilization mode set TPS.
In some embodiments, the processing module 2200 is configured to optimize the power consumption mode of the end user according to the time-of-use power rate structure and the power rate level by using a time-of-use power rate constraint model, a power demand elasticity constraint model, and a distribution network operation constraint model;
wherein the time-of-use electricity price constraint model is as follows:
y i,j =(y i,1,j ,y i,2,j ,…,y i,t,j ,…,y i,|T|,j )
wherein, N pb A step number representing a time of use electricity price for the retail price package; y is i,t,j Is a Boolean variable, whose value is 1 or 0, which indicates whether the ith electricity price block covers t in the day, if so, y i,t,j =1, otherwise y i,t,j =0;Is the electricity purchase cost of the retailer for customer j at i, including costs from forward contracts and spot markets;is the retailer's risk premium for user j.
The elastic constraint model of the power demand is as follows:
wherein r is 0,t Is a nominal retail price; beta is a 0,j ,β 0,j And beta 0,j Are parameters.
The power distribution network operation constraint model is as follows:
U i,j =(u i,1,j ,u i,2,j ,…,u i,|T|,j )
V i,j =(v i,1,j ,v i,2,j ,…,v i,|T|,j )
where ρ is l,k Is the power distribution coefficient, representing the relative change in active power that occurs on line l due to the actual power change at node k;is the power limit of line l; k and L being power distribution networks respectivelyA node set and a line set; u shape i,j And V i,j Are binary vectors whose elements are 0 or 1.
In some embodiments, the execution module 2300 is configured to determine the optimal power usage pattern package using a preset end-user power usage expectation model, a retailer's power purchase cost model, a risk premium model in retail price, and a retailer's expected retail revenue model;
wherein the end user's power usage expectation model is:
wherein, the first and the second end of the pipe are connected with each other,the t-th element of the power usage pattern of end user j.
The electricity purchase cost model of the retailer is as follows:
L j,t =n(E(L j,t ))·E(Q j )
wherein, N F Is the number of forward contracts the retailer has signed;andrespectively indicating the number and price level of the retailer's contracted forward contracts m; beta is a fc Is a weighted factor, β, between the expected revenue and profit risk of the retailer fc ∈[0,+∞);Representing retail risks arising from differences between forward contracts and end-user load expectations; α represents a value-at-risk (VaR) for calculating a conditional value-at-risk (CVaR); β is a given confidence level; n is a radical of hydrogen S Represents the number of samples;is the cost of electricity purchased by the retailer from the spot market;is a binary variable, ifRepresenting the lead time belonging to the forward contract m at t;is the spot price at t; e (L) j,t ) An expectation of electricity usage by the time end user j at time t; e (Q) j ) Indicating the end user j's desire for daily electricity usage.
A risk premium model in the retail price:
the expected retail revenue model for the retailer:
wherein W represents the retail profit of the retailer; e represents the profitability of the retailer's demand.
In some embodiments, the processing module 1200 is further configured to linearize retail risk due to a difference between the forward contract and the end-user load expected value:
M j,n ≥0
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by a computer program, which may be stored in a computer readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. The storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless explicitly stated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
The foregoing is only a partial embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (8)
1. A power package recommendation method based on load curve clustering analysis is characterized by comprising the following steps:
acquiring historical electricity utilization data of terminal users, and analyzing a typical electricity utilization mode of each terminal user by using a DBSCAN clustering algorithm;
optimizing the combination in the typical electricity utilization mode of the terminal user according to a time-of-use electricity price structure and an electricity price level by using a mixed integer nonlinear programming method, and determining an optimal electricity utilization mode package;
recommending the optimal power consumption mode package for each terminal user to the terminal user;
wherein, the electricity utilization mode of the end user is calculated by utilizing a DBSCAN clustering algorithm formulaComprises the following steps:
wherein the vectorRepresenting the end userIn thatThe electricity consumption vector of the day is calculated,which represents a standardized operation of the process of the present invention,to (1) aAn element representing the end userIn thatThe first dayThe power consumption of each time period accounts for the ratio of the power consumption of the user on the day and the whole day,a time period representing historical electricity usage data,representing a collection of typical power usage patterns, TP representing a typical power usage pattern for an end user,representing the L2 norm of the vector.
2. The power package recommendation method according to claim 1, wherein the power usage patterns of the end users are optimized according to a time-of-use power rate structure and a power rate level using an optimization objective function of mixed integer nonlinear programming;
wherein, the first and the second end of the pipe are connected with each other,representing a total electricity usage cost for the end user;represents the set of all end users;representing a decision cycle;representing the end userOn the first dayExpectation of electricity consumption in various time periods;representing the end userOn the first dayThe electricity prices of the individual time periods,representing the end userOn the first dayElasticity of electricity demand for each time period with respect to real-time electricity prices.
3. The power package recommendation method of claim 1, wherein said calculating the end user's power usage pattern using a DBSCAN clustering algorithm formula comprises:
Step 2, initializing a core object setInitializing typical number of power modesInitializing an unaccessed data setInitializing a set of typical power usage patterns;
Step 3, forLook upIs/are as follows-set of domain subsamplesIf at allIs-number of samples of domain subsample setSatisfy the requirement ofThen will beAdding a core object set:obtaining a core object set;
step 4, if the core object setThen the average of the normalized end-user historical electricity usage data is added to the set of typical electricity usage patterns:and ending the algorithm;
step 5, if the core object setIn the core object setIn which a core object is randomly selectedInitializing current cluster core object queueUpdating the number of the typical power consumption modesInitializing the current cluster sample setUpdating the set of unaccessed samples;
Step 6, if the current cluster core object queueThen cluster is currently clusteredAfter the generation is finished, the typical electricity utilization mode set is updatedUpdating the core object setTurning to step 5; if the current cluster core object queueThen updating the core object set;
Step 7, in the current cluster core object queueFetching a core objectFind outIs/are as follows-set of domain subsamplesLet us orderUpdating the current cluster sample setUpdating the set of unaccessed samplesUpdateTurning to step 6;
4. The power package recommendation method of claim 2, wherein said optimizing the end user's power usage pattern according to a time of use power rate structure and power rate level using an optimized objective function of mixed integer nonlinear programming comprises:
optimizing the power utilization mode of the terminal user according to the time-of-use power price structure and the power price level by utilizing a time-of-use power price constraint model, a power demand elastic constraint model and a power distribution network operation constraint model;
wherein the time-of-use electricity price constraint model is as follows:
wherein, the first and the second end of the pipe are connected with each other,a step number representing a time of use electricity price for the retail price package;is a Boolean variable having a value of 1 or 0, meaningWhether individual electricity price blocks cover a dayIf it is covered, thenOtherwise;Is the retailer to the userIn thatA cost of electricity purchase in time, the cost including costs from forward contracts and spot markets;is the retailer to the userRisk of premium;
the elastic constraint model of the power demand is as follows:
wherein the content of the first and second substances,is a nominal retail price;,andis a parameter;
the power distribution network operation constraint model is as follows:
wherein, the first and the second end of the pipe are connected with each other,is a power distribution coefficient, indicates that the node isOn the line in response to actual power changesThe relative change in active power that occurs;is a circuitThe power limit of (d);andrespectively a node set and a line set of the power distribution network;andare binary vectors whose elements are 0 or 1.
5. The power package recommendation method of claim 4, wherein the determining an optimal power usage pattern package comprises:
determining the optimal power utilization pattern package by utilizing a preset end user power utilization expectation model, a retailer power purchase cost model, a risk price premium model in the retail price and an expected retail income model of the retailer;
wherein the end user's power usage expectation model is:
wherein the content of the first and second substances,time terminal userIn the power consumption mode ofAn element;
the electricity purchase cost model of the retailer is as follows:
wherein the content of the first and second substances,is the number of forward contracts the retailer has signed;andrespectively indicating to the retailer to make a forward contractQuantity and price level of;is a weighting factor between the expected revenue of the retailer and the profit risk,;representing retail risks arising from differences between forward contracts and end-user load expectations;representing a risk value VaR for calculating a conditional risk value CVaR;is a given confidence level;represents the number of samples;is the cost of electricity purchased by the retailer from the spot market;is a binary variable, ifTo representTime belonging to a long term contractA lead time of (c);is thatSpot price in time;time terminal userIn thatThe desire to use electricity;representing end usersExpectation of daily electricity usage;
a risk premium model in the retail price:
the expected retail revenue model for the retailer:
7. The utility model provides an electric power package recommendation device based on load curve cluster analysis which characterized in that includes:
the system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring historical electricity consumption data of terminal users and analyzing a typical electricity consumption mode of each terminal user by using a DBSCAN clustering algorithm;
the processing module is used for optimizing the combination in the typical electricity utilization mode of the terminal user according to the time-of-use electricity price structure and the electricity price level by using a mixed integer nonlinear programming method, and determining an optimal electricity utilization mode package;
the execution module is used for recommending the optimal power consumption mode package for each terminal user to the terminal user;
wherein the processing module is used for utilizing DBSCANCalculating the electricity utilization mode of the terminal user by a clustering algorithm formulaComprises the following steps:
wherein the vectorRepresenting the end userIn thatThe electricity consumption vector of a day is calculated,it is shown that the operation of the standardization,to (1) aAn element representing the end userIn thatThe first dayThe power consumption of each time section is the ratio of the power consumption of the user on the day,a time period representing historical power usage data,representing a collection of typical power usage patterns, TP representing a typical power usage pattern for an end user,representing the L2 norm of the vector.
8. The power package recommendation device of claim 7,
the execution module is used for optimizing the power utilization mode of the terminal user according to the time-of-use power price structure and the power price level by using an optimization objective function of mixed integer nonlinear programming;
wherein the content of the first and second substances,representing a total electricity cost for the end user;represents the set of all end users;representing a decision period;representing the end userOn the first day(ii) a desire for electricity usage for a time period;representing the end userOn the first dayThe electricity prices of the individual time periods,representing the end userOn the first dayElasticity of electricity demand for each time period with respect to real-time electricity prices.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110730352.2A CN113538041B (en) | 2021-06-29 | 2021-06-29 | Power package recommendation method and device based on load curve clustering analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110730352.2A CN113538041B (en) | 2021-06-29 | 2021-06-29 | Power package recommendation method and device based on load curve clustering analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113538041A CN113538041A (en) | 2021-10-22 |
CN113538041B true CN113538041B (en) | 2022-10-21 |
Family
ID=78097202
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110730352.2A Active CN113538041B (en) | 2021-06-29 | 2021-06-29 | Power package recommendation method and device based on load curve clustering analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113538041B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116596640A (en) * | 2023-07-19 | 2023-08-15 | 国网山东省电力公司营销服务中心(计量中心) | Recommendation method, system, equipment and storage medium for electric power retail electric charge package |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109086922A (en) * | 2018-07-19 | 2018-12-25 | 浙江大学 | A kind of demand response electric power set meal optimum design method towards industry and commerce user |
CN109325608A (en) * | 2018-06-01 | 2019-02-12 | 国网上海市电力公司 | Consider the distributed generation resource Optimal Configuration Method of energy storage and meter and photovoltaic randomness |
CN110796283A (en) * | 2019-09-18 | 2020-02-14 | 浙江大学 | Demand side active response oriented electric quantity package optimization design method |
CN111242458A (en) * | 2020-01-07 | 2020-06-05 | 广东电网有限责任公司电力调度控制中心 | Electric power retail pricing method and device based on personalized power demand |
CN112200596A (en) * | 2020-09-01 | 2021-01-08 | 中国南方电网有限责任公司 | Method, system, device and medium for determining regional marginal electricity price of power system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11055732B2 (en) * | 2018-09-12 | 2021-07-06 | Mitsubishi Electric Research Laboratories, Inc. | Day-ahead joint generation scheduling and bidding optimization for power plants |
-
2021
- 2021-06-29 CN CN202110730352.2A patent/CN113538041B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109325608A (en) * | 2018-06-01 | 2019-02-12 | 国网上海市电力公司 | Consider the distributed generation resource Optimal Configuration Method of energy storage and meter and photovoltaic randomness |
CN109086922A (en) * | 2018-07-19 | 2018-12-25 | 浙江大学 | A kind of demand response electric power set meal optimum design method towards industry and commerce user |
CN110796283A (en) * | 2019-09-18 | 2020-02-14 | 浙江大学 | Demand side active response oriented electric quantity package optimization design method |
CN111242458A (en) * | 2020-01-07 | 2020-06-05 | 广东电网有限责任公司电力调度控制中心 | Electric power retail pricing method and device based on personalized power demand |
CN112200596A (en) * | 2020-09-01 | 2021-01-08 | 中国南方电网有限责任公司 | Method, system, device and medium for determining regional marginal electricity price of power system |
Non-Patent Citations (1)
Title |
---|
电力现货市场中标准零售套餐设计――基于用户分群的分析;吴敬慧等;《价格理论与实践》;20200306(第12期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113538041A (en) | 2021-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200372588A1 (en) | Methods and systems for machine-learning for prediction of grid carbon emissions | |
US10430898B2 (en) | Method and system for facilitating electricity services | |
CN108711077B (en) | Photovoltaic type microgrid transaction method based on block chain technology | |
US8818889B2 (en) | Technique for aggregating an energy service | |
US9235847B2 (en) | Energy-disutility modeling for agile demand response | |
US20120078687A1 (en) | System and method for lowest cost aggregate energy demand reduction | |
Zhang et al. | Structure-aware stochastic storage management in smart grids | |
Van Ackooij et al. | Optimizing power generation in the presence of micro-grids | |
WO2013102932A2 (en) | System and method facilitating forecasting, optimization and visualization of energy data for an industry | |
CN110852519A (en) | Optimal profit method considering various types of loads for electricity selling companies | |
Yang et al. | Large-Scale Home Energy Management Using Entropy-Based Collective Multiagent Deep Reinforcement Learning Framework. | |
CN109784594B (en) | Adjustable load decision method and system for electronic vendor | |
Konda et al. | Impact of load profile on dynamic interactions between energy markets: a case study of power exchange and demand response exchange | |
Joo et al. | Option valuation applied to implementing demand response via critical peak pricing | |
Finnah et al. | Optimizing trading decisions of wind power plants with hybrid energy storage systems using backwards approximate dynamic programming | |
CN113538041B (en) | Power package recommendation method and device based on load curve clustering analysis | |
Herre et al. | Impact of market timing on the profit of a risk-averse load aggregator | |
CN114757710A (en) | Evaluation method for load aggregator bidding | |
Afzaal et al. | Agent-based energy consumption scheduling for smart grids: an auction-theoretic approach | |
Nammouchi et al. | Robust opportunistic optimal energy management of a mixed microgrid under asymmetrical uncertainties | |
Zheng et al. | An application of machine learning for a smart grid resource allocation problem | |
Shavandi et al. | Extended opportunity cost model to find near equilibrium electricity prices under non-convexities | |
Jia et al. | An online learning approach to dynamic pricing for demand response | |
CN112633771B (en) | Power resource numerical value calculation method, device, computer equipment and storage medium | |
Tesser et al. | An oligopoly model for medium-term power planning in a liberalized electricity market |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |